ORIGINAL ARTICLE Changing trends in blood transfusion: an analysis of 244,013 hospitalizations Nadine Shehata,1,2,3 Alan Forster,4,5 Nadine Lawrence,4 Deanna M. Rothwell,4 Dean Fergusson,4 Alan Tinmouth,4,5 and Kumanan Wilson4,5
BACKGROUND: Identifying recipients of blood transfusion and the trends in transfusion are needed to properly identify and target clinical services in need of patient blood management strategies. We determined the proportion of admissions to each clinical service that received blood, the mean number of units utilized, and the 5-year trends in utilization. STUDY DESIGN AND METHODS: We used a large administrative database, a repository for three campuses of one university-affiliated hospital, and included all adults that were hospitalized from November 1, 2006, to June 2012. The data were analyzed as the proportion of admissions transfused and the mean number units transfused per admission. RESULTS: Of 244,013 hospitalizations, 38,265 received at least one transfusion (29,165 for red blood cells [RBCs], 6760 for plasma, and 5795 for platelets [PLTs]). Although there has been a gradual decrease in the mean number of RBCs transfused (percent change, −9.8%; p = 0.002), an increase in the proportion of admissions receiving RBCs (17.2% increase, p < 0.0001) and PLTs (31.5% increase, p < 0.0001) was apparent while there has been a decrease in the proportion of admissions receiving plasma (23.9% decrease, p < 0.0001). Eight percent of cardiology admissions received RBCs, and the highest mean RBC utilization per admission, aside from the stem cell transplantation service, occurred in cardiology and critical care hospitalizations (mean, 4.7 units/hospitalization). CONCLUSION: Although there has been a reduction in the mean RBC units used, there has been an increase in the proportion of hospitalizations transfused. A better understanding of the indications for transfusion is required to facilitate the development of targeted blood conservation strategies.
T
ransfusion therapy is used for several medical conditions, yet there are limited populationbased data, aside from surveys, on trends in transfusion.1-4 Determining the prevalence and trends of transfusion across patient populations prioritizes research initiatives, identifies patient groups who may benefit from alternatives to transfusion, and predicts the demand for transfusion.2 We categorized the distribution of transfusion in three campuses of a universityaffiliated hospital according to admitting service and assessed the trends in transfusion therapy over a 5-year period with the intent of identifying services with high blood product utilization and a high proportion of patients transfused that may benefit from blood conservation methods.
ABBREVIATION: OHDW = Ottawa Hospital’s Data Warehouse. From the 1Departments of Medicine and Laboratory Medicine and Pathology, Mount Sinai Hospital, Institute of Health Policy Management and Evaluation, University of Toronto; the 2Li Ka Shing Knowledge Institute, St Michael’s Hospital; and the 3 Central Ontario Region, Canadian Blood Services, Toronto, Ontario, Canada; and the 4Ottawa Hospital Research Institute; and the 5Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada. Address reprint requests to: Nadine Shehata, MD, Mount Sinai Hospital, 6-502-5, 600 University Avenue Toronto, ON, Canada M5G 1X5; e-mail: [email protected]. NS is a consultant for Canadian Blood Services. Canadian Blood Services did not have a role in the design, analysis, or interpretation of the results. Funded by a Canadian Blood Services’ intramural grant XT0085, Canada. NS received support from the Canadian Institute of Health Research/Canadian Blood Services New Investigator Award. Received for publication September 4, 2013; revision received January 6, 2014, and accepted January 6, 2014. doi: 10.1111/trf.12644 © 2014 AABB TRANSFUSION **;**:**-**. Volume **, ** **
TRANSFUSION
1
SHEHATA ET AL.
MATERIALS AND METHODS We used information from the Ottawa Hospital’s Data Warehouse (OHDW) in Ottawa, Ontario, Canada. The OHDW is a data repository from the hospital’s operational information systems. These systems include patient registration, the discharge abstract database, laboratory, pharmacy, and radiology results from 2006 onward. This study was approved by the research ethics board at the University of Ottawa. All adult patients that were hospitalized at the three campuses of the university-affiliated Ottawa Hospital from November 1, 2006, to June 2012 were included. The Ottawa Hospital serves 1.2 million residents of Ottawa and Eastern Ontario. We used the OHDW to obtain data on patient characteristics (admitting medical division, emergent vs. nonemergent admission, need for critical care) and blood component transfusion (i.e., plasma, platelets [PLTs], and red blood cells [RBCs]). We categorized transfusion episodes by admitting service and by type of product. We defined admitting service as the first hospital service for each patient hospitalization. The admitting services were categorized as follows: acute care medicine, acute care surgery, anesthesia, stem cell transplantation, cardiac surgery, cardiology, coronary care, dentistry, dermatology, endocrinology, family medicine, general internal medicine, general surgery, geriatrics, gynecology, gynecology/oncology, hematology and oncology, infectious disease, critical care service (admissions to the critical care service included patients in the intensive care unit, patients awaiting intensive care beds but not cardiac patients), nephrology, neurology, neurosurgery, obstetrics, ophthalmology, orthopedics, otorhinolaryngology, plastic surgery, psychiatry, radiation oncology, respirology, thoracic surgery, trauma, urology, and vascular surgery. All gastroenterology admissions were categorized under general internal medicine. We analyzed admissions by transfusable products and subcategorized according to RBCs, plasma, and PLTs. A pooled PLT product was considered as 1 unit and each plasma product was counted as 1 unit. We determined trends over a 5-year interval for all hospitalizations and according to services that had the highest utilization rates for medical and surgical services.
Statistical analyses The data were analyzed as the proportion of hospitalizations receiving blood and the mean number of transfused units per admission receiving blood (e.g., total number of RBCs transfused/total number of admissions who received RBC transfusion). Logistic regression analyses were used to detect significant trends over time for all hospitalizations. Separate run charts were produced to look at the percentage of encounters receiving blood and mean number of units of blood given to encounters 2
TRANSFUSION Volume **, ** **
receiving blood over time (where each time point was 1 month). Similar analyses were conducted for RBC, plasma, and PLT transfusions. The percent change was calculated as 100 × ((intercept + slope × the number of observations) − intercept)/intercept. A p value of less than 0.01 was considered significant.
RESULTS We analyzed 38,265 admissions associated with at least one transfusion among 244,013 hospitalizations. Table 1 illustrates the proportion of hospitalizations using RBCs, plasma, and PLTs utilized by admitting service and Table 2 describes the total number of blood products used by that service. Only services that account for more than 1% of the utilization of these products are displayed. Services that had less than 50 discharges per month are also not displayed (i.e., acute care medicine, anesthesia, coronary care unit, dentistry, dermatology, endocrine, geriatrics, infectious disease, ophthalmology, plastic surgery, psychiatry, radiation oncology, respirology, and trauma).
Usage by service Overall, the general internal medicine service (14%), hematology oncology (14%), and the orthopedic service (13%) accounted for the largest proportions of total admissions receiving RBCs, followed by cardiac surgery (8.8%), general surgery (8.5%), and cardiology (7.6%; Table 1). The cardiac surgery and general internal medicine services accounted for similar proportions of total admissions receiving plasma (16%) with only three other services having proportions higher than 10% (cardiology, general surgery, and the critical care service). The hematology and oncology service (25.2%), cardiac surgery (19.5%), cardiology (12.7%), and stem cell transplantation (11.6%) service accounted for the preponderance of admissions receiving at least one PLT transfusion (Table 1). The services that accounted for more than 10% of the total RBCs used were hematology oncology, general internal medicine, and cardiac surgery (Table 2). Fifteen percent of plasma was used by each of the critical care service and general internal medicine services. Nearly 30% of PLT units were utilized by the hematology oncology service (Table 2). Of all admissions to the stem cell program, 67 and 76% received RBCs and PLTs, respectively, which accounted for the highest proportion of admissions transfused within a service (Table 3). High proportions of admissions to cardiac surgery (35%), hematology oncology (39%), trauma (25%), vascular surgery (24%), and the critical care service received RBCs (29%). The proportions of admissions transfused plasma were evenly distributed
TRENDS IN BLOOD UTILIZATION
Admitting service Cardiac surgery Cardiology Critical care service General internal medicine General surgery Gynecology Gynecology/oncology Hematology and oncology Nephrology Neurosurgery Obstetrics Orthopedics Radiation oncology Stem cell transplantation Urology Vascular surgery
Trends in usage * Data are reported as number (%). Percentages are column percentages. Only services that account for more than 1% of blood products utilized are displayed.
Admissions receiving plasma (total n = 6760) 1047 (15.5) 713 (10.5) 865 (12.8) 1077 (15.9) 792 (11.7) 43 (0.6) 33 (0.5) 377 (5.6) 140 (2.1) 225 (3.3) 66 (1.0) 271 (4.0) 20 (0.3) 54 (0.8) 128 (1.9) 456 (6.7) Admissions receiving RBCs (total n = 29,165) 2563 (8.8) 2211 (7.6) 1570 (5.4) 4095 (14.0) 2477 (8.5) 479 (1.6) 622 (2.1) 4126 (14.1) 700 (2.4) 626 (2.1) 482 (1.7) 3852 (13.2) 483 (1.7) 597 (2.0) 828 (2.8) 1522 (5.2) Admissions receiving any blood product (total n = 38,265) 3034 (7.9) 2541 (6.6) 2410 (6.3) 5351 (14.0) 3170 (8.3) 556 (1.5) 660 (1.7) 4644 (12.1) 857 (2.2) 840 (2.2) 3517 (9.2) 4147 (10.8) 522 (1.4) 751 (2.0) 1003 (2.6) 1666 (4.4) Admissions receiving no blood products (total n = 205,748) 4,255 (2.1) 24,772 (12.0) 3,074 (1.5) 25,021 (12.2) 21,069 (10.2) 7,656 (3.7) 3,105 (1.5) 6,023 (2.9) 3,338 (1.6) 8,836 (4.3) 38,378 (18.7) 19,151 (9.3) 2,074 (1.0) 138 (0.1) 8,861 (4.3) 4,546 (2.2)
TABLE 1. The number of admissions receiving blood products, by admitting service*
Admissions receiving PLTs (total n = 5795) 1130 (19.5) 738 (12.7) 407 (7.0) 363 (6.3) 204 (3.5) 18 (0.3) 22 (0.4) 1462 (25.2) 47 (0.8) 97 (1.7) 52 (0.9) 113 (1.9) 28 (0.5) 675 (11.6) 45 (0.8) 201 (3.5)
among specialties in the range of 1% to 10%, with two services having greater than 10% of hospitalizations transfused (cardiac surgery and the critical care service; Table 3). Aside from the stem cell program, only cardiac surgery and hematology-oncology had more than 10% of admissions receiving PLTs (Table 3).
Figure 1 illustrates the proportion of admissions receiving blood transfusions over 5 years. There has been an increase in the proportion of admissions receiving RBCs (17.2% increase, p < 0.0001) and PLTs (31.5% increase, p < 0.0001), while there has been a decrease in the proportion of admissions receiving plasma (23.9% decrease, p < 0.0001). Figure 2 shows the mean number of units transfused (for the admissions who receive the particular blood product). The number of plasma units transfused per admission has remained constant over time and although the proportion of encounters receiving RBCs and PLTs is increasing, the mean number of units transfused per admission is decreasing. There is a gradual and constant decrease in the number of RBCs transfused (10.9%, p = 0.0002) while the decrease for PLTs (62.7%, p < 0.0001) appears to have tapered since March 2009. Since the proportion of hospitalizations receiving RBCs and PLTs appeared to be increasing but the mean number of units appeared to be decreasing, we examined the overall volume of blood product transfused over time. RBC usage has remained constant and there appears to be a decrease in overall PLT usage until 2009 (Fig. 3). We determined trends in mean RBCs utilized for hospitalizations according to the subgroups of general medicine service and surgical services. Figure 4 displays the four medical services that account for the highest mean number of RBC units utilized aside from the stem cell service (mean, 6.6 units, range, 1-91 units) and hematology and oncology (mean, 4.2 units, range, 1-83 units). The critical care service and cardiology service accounted for the highest mean number of RBC units utilized (mean, 4.8 units, range, 1-92 units; and mean, 4.7 units, range, 1-109 units, respectively). None of the trends displayed a downward tendency, and only the general internal medicine service appears to have a consistent transfusion rate with a mean of 3.2 RBC units (range, 1-64 units). The surgical services displayed more striking fluctuations in the mean number of RBC units transfused particularly for cardiac and vascular surgery with means over 5 years of 4.8 (range, 2.6-8.7) and 3.9 (range, 1.0-13.6) units per admission, respectively (Fig. 5). The urology service had consistent rates.
DISCUSSION We have highlighted that the mean usage of RBCs per hospitalization has decreased but the proportion of Volume **, ** **
TRANSFUSION
3
SHEHATA ET AL.
surgery, critical care, vascular surgery, trauma) whereas the general internal medicine service and cardiac surgery PLT units accounted for the highest proportion of (total n = 24,916) admissions receiving plasma (16%) fol4778 (19.2) lowed by the cardiology service, general 3529 (14.2) 1251 (5.0) surgery, and the critical care service 1220 (4.9) (11%). The percentage of patients 618 (2.5) receiving plasma transfusion is high 41 (0.2) 33 (0.1) despite the lack of effectiveness of the 6901 (27.7) use of plasma in clinical trials and the 109 (0.4) suggestion of potential harm.9 376 (1.5) 97 (0.4) The trend for RBCs (less blood 284 (1.1) administered/admission but more 93 (0.4) admissions receiving blood) is sugges4227 (16.7) 127 (0.5) tive of a shift in RBC use potentially due 749 (3.0) to critical transfusion trials10-13 and the promotion of the concept of patient blood management.14 Yet, the downward trends in the services anticipated to show a decline; for example, the critical care service was not evident. The critical care service exhibited fluctuating rates of RBCs utilized and general internal medicine service showed a constant rate of utilization, albeit high. The general internal medicine service has a heterogeneous patient population, and although the high rate of transfusion of RBCs for the general medical hospitalizations has been previously identified,2 more data are required to accurately describe the patient populations transfused as a better understanding of the rationale for transfusion is required since the concept of patient blood management may be useful for these patients where alternatives to blood transfusion such as iron supplementation in patients who are at risk of receiving transfusion, for example, elderly patients, patients with inflammatory bowel disease, and patients with heart failure, have been shown to be effective.15-19 Additionally, we did not find a downward trend in the number of RBC units utilized per admission for surgical services, cardiology, or nephrology contrary to previous reports,20 and the proportion of admissions receiving RBCs appears to have increased. The basis for this increase similar to the rate for the general internal medicine service needs to be elucidated. The high proportion of admissions receiving RBCs and the high utilization of RBCs for hospitalizations undergoing stem cell transplantation may be reduced if the current pilot feasibility randomized controlled trial of restrictive (70-90 g/L) compared to liberal (90-100 g/L) RBC transfusion strategy favors the restrictive group.21 The hematology-oncology, stem cell transplantation, and cardiac surgery services were the predominant users of PLT transfusion. Whereas for the hematology-oncology and stem cell transplantation services, the indications for PLT transfusion have been well defined,22-24 more data are
TABLE 2. The total number of blood product units used by each admitting service* Admitting service Cardiac surgery Cardiology Critical care service General internal medicine General surgery Gynecology Gynecology/oncology Hematology and oncology Nephrology Neurosurgery Obstetrics Orthopedics Radiation oncology Stem cell transplantation Urology Vascular surgery
RBC units (total n = 111,098) 12,097 (10.9) 10,433 (9.4) 7,625 (6.9) 12,946 (11.7) 9,113 (8.2) 1,307 (1.2) 1,541 (1.4) 17,263 (15.5) 2,109 (1.9) 1,895 (1.7) 1,522 (1.4) 10,950 (9.9) 1,469 (1.3) 3,948 (3.6) 2,828 (2.5) 6,946 (6.3)
Plasma units (total n = 48,542) 6760 (13.9) 5156 (10.6) 7443 (15.3) 7353 (15.2) 4573 (9.4) 237 (0.5) 147 (0.3) 5089 (10.5) 1745 (3.6) 1061 (2.2) 542 (1.1) 1322 (2.7) 88 (0.2) 627 (1.3) 621 (1.3) 3011 (6.2)
* Data are reported as number (%).
hospitalizations receiving RBCs has increased resulting in a trend in overall RBC usage that appears to have remained stable. PLT transfusion has declined over a 5-year interval. The proportion of admissions receiving plasma decreased, but the trend in overall usage has remained relatively constant. The critical care service continues to be a service with a high proportion of admissions receiving RBCs (29%) and plasma (16%), These results are higher than previously published.5,6 We have demonstrated, similar to other studies, that surgical services and services administering chemotherapy (i.e., stem cell transplantation and hematology-oncology) were still associated with the highest rates of RBC use;1-3 yet, we also have identified that close to 20% of nephrology admissions receive RBCs and the general internal medicine service accounts for high proportion of admissions receiving RBCs. Although 8% of cardiology admissions receive RBCs, this service is associated with the highest mean RBC utilization (aside from the stem cell transplantation service; mean, 4.7 units/ admission). This high rate and the fluctuation in RBC utilization may reflect the uncertainty of when to transfuse RBCs because of the lack of definitive trials correlating the degree of anemia with the degree of hypoxemic events7,8 and warrants further study to identify the appropriate hemoglobin thresholds for RBCs. The general internal medicine service and cardiac surgery accounted for equivalent proportions of admissions receiving plasma (approx. 16%), although the proportion of admissions to the general internal medicine service that received plasma (3.5%) was lower than that for the cardiac surgery service (14.4%). Plasma transfusion did not have the same distribution as RBCs; aside from the hematology-oncology and stem cell services, the highest proportions of admissions receiving RBCs (between 25 and 35%) occurred under several services (i.e., cardiac 4
TRANSFUSION Volume **, ** **
TRENDS IN BLOOD UTILIZATION
TABLE 3. The proportions of admissions to each service receiving RBCs, plasma, and PLTs Admitting service Cardiac surgery Cardiology Critical care service Family medicine General internal medicine General surgery Gynecology Gynecology/oncology Hematology and oncology Nephrology Neurosurgery Obstetrics Orthopedics Radiation oncology Stem cell transplantation Trauma Urology Vascular surgery
Total number of admissions 7,289 27,313 5,484 3,997 30,372 24,239 8,212 3,765 10,667 4,195 9,676 41,895 23,298 2,596 889 1,064 9,864 6,212
Admissions receiving RBCs 2563 (35.2) 2211 (8.1) 1570 (28.6) 402 (10.1) 4095 (13.5) 2477 (10.2) 479 (5.8) 622 (16.5) 4126 (38.7) 700 (16.7) 626 (6.4) 482 (1.2) 3852 (16.5) 483 (1.7) 597 (67.1) 267 (25.1) 828 (8.4) 1522 (24.5)
Admissions receiving plasma 1047 (14.4) 713 (2.6) 865 (15.8) 66 (1.6) 1,077 (3.5) 792 (3.3) 43 (0.5) 33 (0.9) 377 (3.5) 140 (3.3) 225 (2.3) 66 (0.1) 271 (1.1) 20 (0.8) 54 (6.1) 100 (9.3) 128 (1.3) 456 (7.3)
Admissions receiving PLTs 1130 (15.5) 738 (2.7) 407 (7.4) 31 (0.8) 363 (1.2) 204 (0.8) 18 (0.2) 22 (0.6) 1462 (13.7) 47 (1.1) 97 (1.0) 52 (0.1) 113 (0.5) 28 (1.1) 675 (75.9) 52 (4.9) 45 (0.8) 201 (0.4)
* Data are reported as number (%). Percentages are row percentages. Only services that account for more than 1% of blood products utilized are displayed.
Fig. 1. The proportion of admissions transfused from November 2006 to June 2012.
Volume **, ** **
TRANSFUSION
5
SHEHATA ET AL.
Fig. 2. The mean number of units transfused per hospitalization for RBCs, plasma, and PLTs from November 2006 to June 2012.
Fig. 3. The number of blood products (units) transfused from November 2006 to June 2012.
6
TRANSFUSION Volume **, ** **
TRENDS IN BLOOD UTILIZATION
Fig. 4. The mean number of RBC units transfused per hospitalization for hospitalizations receiving RBCs, by admitting medical service.
Fig. 5. The mean number of RBC units transfused per hospitalization receiving RBCs by admitting surgical service.
Volume **, ** **
TRANSFUSION
7
SHEHATA ET AL.
still required to identify critical thresholds for transfusion of PLTs in the cardiac surgery population,25 as the proportion of admissions receiving PLT transfusion has increased and it is unclear whether PLT transfusion is associated with increased morbidity and mortality.26,27 There are a few limitations to this study. Studies using administrative data are limited as they cannot describe the indications for transfusion or the use of alternative agents to treat anemia to explain the trends and fluctuations in transfusion. Yet, regardless of the indications, the trends categorized by clinical service identify disciplines where indications need to be further delineated. We did not describe laboratory data or patient characteristics and these features may have impacted on the trends in transfusion such as triggers for transfusion or age, as transfusion increases with older age, and there may have been an increase in the age of captured population in our study over time.28,29 This is a single-institution study and although generalization may be limited, the study included three campuses of a university-affiliated hospital and 244,013 hospitalizations. In summary, a clearer understanding of the indications and laboratory variables used to guide transfusion in patients admitted, particularly under the general internal medicine, nephrology services, and the critical care service for RBCs (as the proportion of admissions transfused in critical care is high despite data demonstrating that lower transfusion thresholds can be used11) is required. The general internal medicine service accounts for a large number of hospital admissions, and limited data are available on the indications for transfusion. Identifying factors that lead to high proportions of patients transfused RBCs under the critical care service may aid in identifying blood conservation strategies to decrease exposure to blood in these populations. Similarly there is the same need to determine the rationale for plasma transfusion in critical care and cardiac surgery and for PLT transfusion in cardiac surgery. The application of administrative data to describe trends in transfusion utilization and the proportion of admissions transfused in this study may also be useful in defining the patient populations and characteristics associated with increased transfusion rates. ACKNOWLEDGMENT We thank Robin Ducharme for her assistance in the preparation of the manuscript. CONFLICT OF INTEREST The authors have disclosed no conflicts of interest.
REFERENCES 1. Chiavetta JA, Herst R, Freedman J, et al. A survey of red cell use in 45 hospitals in central Ontario, Canada. Transfusion 1996;36:699-706. 8
TRANSFUSION Volume **, ** **
2. Wells AW, Mounter PJ, Chapman CE, et al. Where does blood go? Prospective observational study of red cell transfusion in north England. BMJ 2002;325(7368):803. 3. Cobain TJ, Vamvakas EC, Wells A, et al. A survey of the demographics of blood use. Transfus Med 2007;17:1-15. 4. Stanworth SJ, Cockburn HA, Boralessa H, et al. Which groups of patients are transfused? A study of red cell usage in London and southeast England. Vox Sang 2002;83: 352-7. 5. Stanworth SJ, Walsh TS, Prescott RJ, et al.; Intensive Care Study of Coagulopathy (ISOC) investigators. A national study of plasma use in critical care: clinical indications, dose and effect on prothrombin time. Crit Care 2011;15: R108. 6. Walsh TS, Garrioch M, Maciver C, et al.; Audit of Transfusion in Intensive Care in Scotland Study Group. Red cell requirements for intensive care units adhering to evidence-based transfusion guidelines. Transfusion 2004; 44:1405-11. 7. Chatterjee S, Wetterslev J, Sharma A, et al. Association of blood transfusion with increased mortality in myocardial infarction: a meta-analysis and diversity-adjusted study sequential analysis. Arch Intern Med 2013;173:132-9. 8. Carson JL, Hébert PC. Here we go again—blood transfusion kills patients? Comment on “Association of blood transfusion with increased mortality in myocardial infarction: a meta-analysis and diversity-adjusted study sequential analysis.” JAMA Intern Med 2013;173:139-41. 9. Yang L, Stanworth S, Hopewell S, et al. Is fresh-frozen plasma clinically effective? An update of a systematic review of randomized controlled trials. Transfusion 2012; 52:1673-86; quiz 1673. 10. Corwin HL, Gettinger A, Pearl RG, et al. The CRIT Study: anemia and blood transfusion in the critically ill—current clinical practice in the United States. Crit Care Med 2004; 32:39-52. 11. Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999;340:409-17. 12. Hébert PC, Blajchman MA, Cook DJ, et al.; Transfusion Requirements in Critical Care Investigators for the Canadian Critical Care Trials Group. Do blood transfusions improve outcomes related to mechanical ventilation? Chest 2001;119:1850-7. 13. Vincent JL, Baron JF, Reinhart K, et al. Anemia and blood transfusion in critically ill patients. JAMA 2002;288:1499507. 14. Williamson LM, Devine DV. Challenges in the management of the blood supply. Lancet 2013;381(9880):1866-75. 15. Guralnik JM, Eisenstaedt RS, Ferrucci L, et al. Prevalence of anemia in persons 65 years and older in the United States: evidence for a high rate of unexplained anemia. Blood 2004;104:2263-8.
TRENDS IN BLOOD UTILIZATION
16. Kulnigg S, Gasche C. Systematic review: managing anaemia in Crohn’s disease. Aliment Pharmacol Ther 2006;24(1112):1507-23. 17. Okonko DO, Grzeslo A, Witkowski T, et al. Effect of intravenous iron sucrose on exercise tolerance in anemic and nonanemic patients with symptomatic chronic heart failure and iron deficiency FERRIC-HF: a randomized, controlled, observer-blinded trial. J Am Coll Cardiol 2008;51: 103-12. 18. Silverberg DS, Iaina A, Schwartz D, et al. Intravenous iron in heart failure: beyond targeting anemia. Curr Heart Fail Rep 2011;8:14-21. 19. Auerbach M, Goodnough LT, Picard D, et al. The role of intravenous iron in anemia management and transfusion avoidance. Transfusion 2008;48:988-1000. 20. Wallis JP, Wells AW, Chapman CE. Changing indications
23. Estcourt L, Stanworth S, Doree C, et al. Prophylactic platelet transfusion for prevention of bleeding in patients with haematological disorders after chemotherapy and stem cell transplantation. Cochrane Database Syst Rev 2012;(5): CD004269. 24. Wandt H, Schaefer-Eckart K, Wendelin K, et al.; Study Alliance Leukemia. Therapeutic platelet transfusion versus routine prophylactic transfusion in patients with haematological malignancies: an open-label, multicentre, randomised study. Lancet 2012;380(9850):1309-16. 25. Zaffar N, Joseph A, Mazer CD, et al. The rationale for platelet transfusion during cardiopulmonary bypass: an observational study. Can J Anaesth 2013;60:345-54. 26. Spiess BD, Royston D, Levy JH, et al. Platelet transfusions during coronary artery bypass graft surgery are associated with serious adverse outcomes. Transfusion 2004;44:
for red cell transfusion from 2000 to 2004 in the North of England. Transfus Med 2006;16:411-7.
1143-8. 27. Karkouti K, Wijeysundera DN, Yau TM, et al. Platelet trans-
21. Tay J, Tinmouth A, Fergusson D, et al. Transfusion of red cells in hematopoietic stem cell transplantation (TRIST): study protocol for a randomized controlled trial. Trials
fusions are not associated with increased morbidity or mortality in cardiac surgery. Can J Anaesth 2006;53:279-87. 28. Drackley A, Newbold KB, Paez A, et al. Forecasting Ontar-
2011;12:207. 22. Stanworth SJ, Estcourt LJ, Powter G, et al.; TOPPS Investigators. A no-prophylaxis platelet-transfusion strategy for hematologic cancers. N Engl J Med 2013;368: 1771-80.
io’s blood supply and demand. Transfusion 2012;52:36674. 29. Greinacher A, Fendrich K, Brzenska R, et al. Implications of demographics on future blood supply: a population-based cross-sectional study. Transfusion 2011;51:702-9.
Volume **, ** **
TRANSFUSION
9
BACKGROUND: Identifying recipients of blood transfusion and the trends in transfusion are needed to properly identify and target clinical services in need of patient blood management strategies. We determined the proportion of admissions to each clinical service that received blood, the mean number of units utilized, and the 5-year trends in utilization. STUDY DESIGN AND METHODS: We used a large administrative database, a repository for three campuses of one university-affiliated hospital, and included all adults that were hospitalized from November 1, 2006, to June 2012. The data were analyzed as the proportion of admissions transfused and the mean number units transfused per admission. RESULTS: Of 244,013 hospitalizations, 38,265 received at least one transfusion (29,165 for red blood cells [RBCs], 6760 for plasma, and 5795 for platelets [PLTs]). Although there has been a gradual decrease in the mean number of RBCs transfused (percent change, −9.8%; p = 0.002), an increase in the proportion of admissions receiving RBCs (17.2% increase, p < 0.0001) and PLTs (31.5% increase, p < 0.0001) was apparent while there has been a decrease in the proportion of admissions receiving plasma (23.9% decrease, p < 0.0001). Eight percent of cardiology admissions received RBCs, and the highest mean RBC utilization per admission, aside from the stem cell transplantation service, occurred in cardiology and critical care hospitalizations (mean, 4.7 units/hospitalization). CONCLUSION: Although there has been a reduction in the mean RBC units used, there has been an increase in the proportion of hospitalizations transfused. A better understanding of the indications for transfusion is required to facilitate the development of targeted blood conservation strategies.
T
ransfusion therapy is used for several medical conditions, yet there are limited populationbased data, aside from surveys, on trends in transfusion.1-4 Determining the prevalence and trends of transfusion across patient populations prioritizes research initiatives, identifies patient groups who may benefit from alternatives to transfusion, and predicts the demand for transfusion.2 We categorized the distribution of transfusion in three campuses of a universityaffiliated hospital according to admitting service and assessed the trends in transfusion therapy over a 5-year period with the intent of identifying services with high blood product utilization and a high proportion of patients transfused that may benefit from blood conservation methods.
ABBREVIATION: OHDW = Ottawa Hospital’s Data Warehouse. From the 1Departments of Medicine and Laboratory Medicine and Pathology, Mount Sinai Hospital, Institute of Health Policy Management and Evaluation, University of Toronto; the 2Li Ka Shing Knowledge Institute, St Michael’s Hospital; and the 3 Central Ontario Region, Canadian Blood Services, Toronto, Ontario, Canada; and the 4Ottawa Hospital Research Institute; and the 5Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada. Address reprint requests to: Nadine Shehata, MD, Mount Sinai Hospital, 6-502-5, 600 University Avenue Toronto, ON, Canada M5G 1X5; e-mail: [email protected]. NS is a consultant for Canadian Blood Services. Canadian Blood Services did not have a role in the design, analysis, or interpretation of the results. Funded by a Canadian Blood Services’ intramural grant XT0085, Canada. NS received support from the Canadian Institute of Health Research/Canadian Blood Services New Investigator Award. Received for publication September 4, 2013; revision received January 6, 2014, and accepted January 6, 2014. doi: 10.1111/trf.12644 © 2014 AABB TRANSFUSION **;**:**-**. Volume **, ** **
TRANSFUSION
1
SHEHATA ET AL.
MATERIALS AND METHODS We used information from the Ottawa Hospital’s Data Warehouse (OHDW) in Ottawa, Ontario, Canada. The OHDW is a data repository from the hospital’s operational information systems. These systems include patient registration, the discharge abstract database, laboratory, pharmacy, and radiology results from 2006 onward. This study was approved by the research ethics board at the University of Ottawa. All adult patients that were hospitalized at the three campuses of the university-affiliated Ottawa Hospital from November 1, 2006, to June 2012 were included. The Ottawa Hospital serves 1.2 million residents of Ottawa and Eastern Ontario. We used the OHDW to obtain data on patient characteristics (admitting medical division, emergent vs. nonemergent admission, need for critical care) and blood component transfusion (i.e., plasma, platelets [PLTs], and red blood cells [RBCs]). We categorized transfusion episodes by admitting service and by type of product. We defined admitting service as the first hospital service for each patient hospitalization. The admitting services were categorized as follows: acute care medicine, acute care surgery, anesthesia, stem cell transplantation, cardiac surgery, cardiology, coronary care, dentistry, dermatology, endocrinology, family medicine, general internal medicine, general surgery, geriatrics, gynecology, gynecology/oncology, hematology and oncology, infectious disease, critical care service (admissions to the critical care service included patients in the intensive care unit, patients awaiting intensive care beds but not cardiac patients), nephrology, neurology, neurosurgery, obstetrics, ophthalmology, orthopedics, otorhinolaryngology, plastic surgery, psychiatry, radiation oncology, respirology, thoracic surgery, trauma, urology, and vascular surgery. All gastroenterology admissions were categorized under general internal medicine. We analyzed admissions by transfusable products and subcategorized according to RBCs, plasma, and PLTs. A pooled PLT product was considered as 1 unit and each plasma product was counted as 1 unit. We determined trends over a 5-year interval for all hospitalizations and according to services that had the highest utilization rates for medical and surgical services.
Statistical analyses The data were analyzed as the proportion of hospitalizations receiving blood and the mean number of transfused units per admission receiving blood (e.g., total number of RBCs transfused/total number of admissions who received RBC transfusion). Logistic regression analyses were used to detect significant trends over time for all hospitalizations. Separate run charts were produced to look at the percentage of encounters receiving blood and mean number of units of blood given to encounters 2
TRANSFUSION Volume **, ** **
receiving blood over time (where each time point was 1 month). Similar analyses were conducted for RBC, plasma, and PLT transfusions. The percent change was calculated as 100 × ((intercept + slope × the number of observations) − intercept)/intercept. A p value of less than 0.01 was considered significant.
RESULTS We analyzed 38,265 admissions associated with at least one transfusion among 244,013 hospitalizations. Table 1 illustrates the proportion of hospitalizations using RBCs, plasma, and PLTs utilized by admitting service and Table 2 describes the total number of blood products used by that service. Only services that account for more than 1% of the utilization of these products are displayed. Services that had less than 50 discharges per month are also not displayed (i.e., acute care medicine, anesthesia, coronary care unit, dentistry, dermatology, endocrine, geriatrics, infectious disease, ophthalmology, plastic surgery, psychiatry, radiation oncology, respirology, and trauma).
Usage by service Overall, the general internal medicine service (14%), hematology oncology (14%), and the orthopedic service (13%) accounted for the largest proportions of total admissions receiving RBCs, followed by cardiac surgery (8.8%), general surgery (8.5%), and cardiology (7.6%; Table 1). The cardiac surgery and general internal medicine services accounted for similar proportions of total admissions receiving plasma (16%) with only three other services having proportions higher than 10% (cardiology, general surgery, and the critical care service). The hematology and oncology service (25.2%), cardiac surgery (19.5%), cardiology (12.7%), and stem cell transplantation (11.6%) service accounted for the preponderance of admissions receiving at least one PLT transfusion (Table 1). The services that accounted for more than 10% of the total RBCs used were hematology oncology, general internal medicine, and cardiac surgery (Table 2). Fifteen percent of plasma was used by each of the critical care service and general internal medicine services. Nearly 30% of PLT units were utilized by the hematology oncology service (Table 2). Of all admissions to the stem cell program, 67 and 76% received RBCs and PLTs, respectively, which accounted for the highest proportion of admissions transfused within a service (Table 3). High proportions of admissions to cardiac surgery (35%), hematology oncology (39%), trauma (25%), vascular surgery (24%), and the critical care service received RBCs (29%). The proportions of admissions transfused plasma were evenly distributed
TRENDS IN BLOOD UTILIZATION
Admitting service Cardiac surgery Cardiology Critical care service General internal medicine General surgery Gynecology Gynecology/oncology Hematology and oncology Nephrology Neurosurgery Obstetrics Orthopedics Radiation oncology Stem cell transplantation Urology Vascular surgery
Trends in usage * Data are reported as number (%). Percentages are column percentages. Only services that account for more than 1% of blood products utilized are displayed.
Admissions receiving plasma (total n = 6760) 1047 (15.5) 713 (10.5) 865 (12.8) 1077 (15.9) 792 (11.7) 43 (0.6) 33 (0.5) 377 (5.6) 140 (2.1) 225 (3.3) 66 (1.0) 271 (4.0) 20 (0.3) 54 (0.8) 128 (1.9) 456 (6.7) Admissions receiving RBCs (total n = 29,165) 2563 (8.8) 2211 (7.6) 1570 (5.4) 4095 (14.0) 2477 (8.5) 479 (1.6) 622 (2.1) 4126 (14.1) 700 (2.4) 626 (2.1) 482 (1.7) 3852 (13.2) 483 (1.7) 597 (2.0) 828 (2.8) 1522 (5.2) Admissions receiving any blood product (total n = 38,265) 3034 (7.9) 2541 (6.6) 2410 (6.3) 5351 (14.0) 3170 (8.3) 556 (1.5) 660 (1.7) 4644 (12.1) 857 (2.2) 840 (2.2) 3517 (9.2) 4147 (10.8) 522 (1.4) 751 (2.0) 1003 (2.6) 1666 (4.4) Admissions receiving no blood products (total n = 205,748) 4,255 (2.1) 24,772 (12.0) 3,074 (1.5) 25,021 (12.2) 21,069 (10.2) 7,656 (3.7) 3,105 (1.5) 6,023 (2.9) 3,338 (1.6) 8,836 (4.3) 38,378 (18.7) 19,151 (9.3) 2,074 (1.0) 138 (0.1) 8,861 (4.3) 4,546 (2.2)
TABLE 1. The number of admissions receiving blood products, by admitting service*
Admissions receiving PLTs (total n = 5795) 1130 (19.5) 738 (12.7) 407 (7.0) 363 (6.3) 204 (3.5) 18 (0.3) 22 (0.4) 1462 (25.2) 47 (0.8) 97 (1.7) 52 (0.9) 113 (1.9) 28 (0.5) 675 (11.6) 45 (0.8) 201 (3.5)
among specialties in the range of 1% to 10%, with two services having greater than 10% of hospitalizations transfused (cardiac surgery and the critical care service; Table 3). Aside from the stem cell program, only cardiac surgery and hematology-oncology had more than 10% of admissions receiving PLTs (Table 3).
Figure 1 illustrates the proportion of admissions receiving blood transfusions over 5 years. There has been an increase in the proportion of admissions receiving RBCs (17.2% increase, p < 0.0001) and PLTs (31.5% increase, p < 0.0001), while there has been a decrease in the proportion of admissions receiving plasma (23.9% decrease, p < 0.0001). Figure 2 shows the mean number of units transfused (for the admissions who receive the particular blood product). The number of plasma units transfused per admission has remained constant over time and although the proportion of encounters receiving RBCs and PLTs is increasing, the mean number of units transfused per admission is decreasing. There is a gradual and constant decrease in the number of RBCs transfused (10.9%, p = 0.0002) while the decrease for PLTs (62.7%, p < 0.0001) appears to have tapered since March 2009. Since the proportion of hospitalizations receiving RBCs and PLTs appeared to be increasing but the mean number of units appeared to be decreasing, we examined the overall volume of blood product transfused over time. RBC usage has remained constant and there appears to be a decrease in overall PLT usage until 2009 (Fig. 3). We determined trends in mean RBCs utilized for hospitalizations according to the subgroups of general medicine service and surgical services. Figure 4 displays the four medical services that account for the highest mean number of RBC units utilized aside from the stem cell service (mean, 6.6 units, range, 1-91 units) and hematology and oncology (mean, 4.2 units, range, 1-83 units). The critical care service and cardiology service accounted for the highest mean number of RBC units utilized (mean, 4.8 units, range, 1-92 units; and mean, 4.7 units, range, 1-109 units, respectively). None of the trends displayed a downward tendency, and only the general internal medicine service appears to have a consistent transfusion rate with a mean of 3.2 RBC units (range, 1-64 units). The surgical services displayed more striking fluctuations in the mean number of RBC units transfused particularly for cardiac and vascular surgery with means over 5 years of 4.8 (range, 2.6-8.7) and 3.9 (range, 1.0-13.6) units per admission, respectively (Fig. 5). The urology service had consistent rates.
DISCUSSION We have highlighted that the mean usage of RBCs per hospitalization has decreased but the proportion of Volume **, ** **
TRANSFUSION
3
SHEHATA ET AL.
surgery, critical care, vascular surgery, trauma) whereas the general internal medicine service and cardiac surgery PLT units accounted for the highest proportion of (total n = 24,916) admissions receiving plasma (16%) fol4778 (19.2) lowed by the cardiology service, general 3529 (14.2) 1251 (5.0) surgery, and the critical care service 1220 (4.9) (11%). The percentage of patients 618 (2.5) receiving plasma transfusion is high 41 (0.2) 33 (0.1) despite the lack of effectiveness of the 6901 (27.7) use of plasma in clinical trials and the 109 (0.4) suggestion of potential harm.9 376 (1.5) 97 (0.4) The trend for RBCs (less blood 284 (1.1) administered/admission but more 93 (0.4) admissions receiving blood) is sugges4227 (16.7) 127 (0.5) tive of a shift in RBC use potentially due 749 (3.0) to critical transfusion trials10-13 and the promotion of the concept of patient blood management.14 Yet, the downward trends in the services anticipated to show a decline; for example, the critical care service was not evident. The critical care service exhibited fluctuating rates of RBCs utilized and general internal medicine service showed a constant rate of utilization, albeit high. The general internal medicine service has a heterogeneous patient population, and although the high rate of transfusion of RBCs for the general medical hospitalizations has been previously identified,2 more data are required to accurately describe the patient populations transfused as a better understanding of the rationale for transfusion is required since the concept of patient blood management may be useful for these patients where alternatives to blood transfusion such as iron supplementation in patients who are at risk of receiving transfusion, for example, elderly patients, patients with inflammatory bowel disease, and patients with heart failure, have been shown to be effective.15-19 Additionally, we did not find a downward trend in the number of RBC units utilized per admission for surgical services, cardiology, or nephrology contrary to previous reports,20 and the proportion of admissions receiving RBCs appears to have increased. The basis for this increase similar to the rate for the general internal medicine service needs to be elucidated. The high proportion of admissions receiving RBCs and the high utilization of RBCs for hospitalizations undergoing stem cell transplantation may be reduced if the current pilot feasibility randomized controlled trial of restrictive (70-90 g/L) compared to liberal (90-100 g/L) RBC transfusion strategy favors the restrictive group.21 The hematology-oncology, stem cell transplantation, and cardiac surgery services were the predominant users of PLT transfusion. Whereas for the hematology-oncology and stem cell transplantation services, the indications for PLT transfusion have been well defined,22-24 more data are
TABLE 2. The total number of blood product units used by each admitting service* Admitting service Cardiac surgery Cardiology Critical care service General internal medicine General surgery Gynecology Gynecology/oncology Hematology and oncology Nephrology Neurosurgery Obstetrics Orthopedics Radiation oncology Stem cell transplantation Urology Vascular surgery
RBC units (total n = 111,098) 12,097 (10.9) 10,433 (9.4) 7,625 (6.9) 12,946 (11.7) 9,113 (8.2) 1,307 (1.2) 1,541 (1.4) 17,263 (15.5) 2,109 (1.9) 1,895 (1.7) 1,522 (1.4) 10,950 (9.9) 1,469 (1.3) 3,948 (3.6) 2,828 (2.5) 6,946 (6.3)
Plasma units (total n = 48,542) 6760 (13.9) 5156 (10.6) 7443 (15.3) 7353 (15.2) 4573 (9.4) 237 (0.5) 147 (0.3) 5089 (10.5) 1745 (3.6) 1061 (2.2) 542 (1.1) 1322 (2.7) 88 (0.2) 627 (1.3) 621 (1.3) 3011 (6.2)
* Data are reported as number (%).
hospitalizations receiving RBCs has increased resulting in a trend in overall RBC usage that appears to have remained stable. PLT transfusion has declined over a 5-year interval. The proportion of admissions receiving plasma decreased, but the trend in overall usage has remained relatively constant. The critical care service continues to be a service with a high proportion of admissions receiving RBCs (29%) and plasma (16%), These results are higher than previously published.5,6 We have demonstrated, similar to other studies, that surgical services and services administering chemotherapy (i.e., stem cell transplantation and hematology-oncology) were still associated with the highest rates of RBC use;1-3 yet, we also have identified that close to 20% of nephrology admissions receive RBCs and the general internal medicine service accounts for high proportion of admissions receiving RBCs. Although 8% of cardiology admissions receive RBCs, this service is associated with the highest mean RBC utilization (aside from the stem cell transplantation service; mean, 4.7 units/ admission). This high rate and the fluctuation in RBC utilization may reflect the uncertainty of when to transfuse RBCs because of the lack of definitive trials correlating the degree of anemia with the degree of hypoxemic events7,8 and warrants further study to identify the appropriate hemoglobin thresholds for RBCs. The general internal medicine service and cardiac surgery accounted for equivalent proportions of admissions receiving plasma (approx. 16%), although the proportion of admissions to the general internal medicine service that received plasma (3.5%) was lower than that for the cardiac surgery service (14.4%). Plasma transfusion did not have the same distribution as RBCs; aside from the hematology-oncology and stem cell services, the highest proportions of admissions receiving RBCs (between 25 and 35%) occurred under several services (i.e., cardiac 4
TRANSFUSION Volume **, ** **
TRENDS IN BLOOD UTILIZATION
TABLE 3. The proportions of admissions to each service receiving RBCs, plasma, and PLTs Admitting service Cardiac surgery Cardiology Critical care service Family medicine General internal medicine General surgery Gynecology Gynecology/oncology Hematology and oncology Nephrology Neurosurgery Obstetrics Orthopedics Radiation oncology Stem cell transplantation Trauma Urology Vascular surgery
Total number of admissions 7,289 27,313 5,484 3,997 30,372 24,239 8,212 3,765 10,667 4,195 9,676 41,895 23,298 2,596 889 1,064 9,864 6,212
Admissions receiving RBCs 2563 (35.2) 2211 (8.1) 1570 (28.6) 402 (10.1) 4095 (13.5) 2477 (10.2) 479 (5.8) 622 (16.5) 4126 (38.7) 700 (16.7) 626 (6.4) 482 (1.2) 3852 (16.5) 483 (1.7) 597 (67.1) 267 (25.1) 828 (8.4) 1522 (24.5)
Admissions receiving plasma 1047 (14.4) 713 (2.6) 865 (15.8) 66 (1.6) 1,077 (3.5) 792 (3.3) 43 (0.5) 33 (0.9) 377 (3.5) 140 (3.3) 225 (2.3) 66 (0.1) 271 (1.1) 20 (0.8) 54 (6.1) 100 (9.3) 128 (1.3) 456 (7.3)
Admissions receiving PLTs 1130 (15.5) 738 (2.7) 407 (7.4) 31 (0.8) 363 (1.2) 204 (0.8) 18 (0.2) 22 (0.6) 1462 (13.7) 47 (1.1) 97 (1.0) 52 (0.1) 113 (0.5) 28 (1.1) 675 (75.9) 52 (4.9) 45 (0.8) 201 (0.4)
* Data are reported as number (%). Percentages are row percentages. Only services that account for more than 1% of blood products utilized are displayed.
Fig. 1. The proportion of admissions transfused from November 2006 to June 2012.
Volume **, ** **
TRANSFUSION
5
SHEHATA ET AL.
Fig. 2. The mean number of units transfused per hospitalization for RBCs, plasma, and PLTs from November 2006 to June 2012.
Fig. 3. The number of blood products (units) transfused from November 2006 to June 2012.
6
TRANSFUSION Volume **, ** **
TRENDS IN BLOOD UTILIZATION
Fig. 4. The mean number of RBC units transfused per hospitalization for hospitalizations receiving RBCs, by admitting medical service.
Fig. 5. The mean number of RBC units transfused per hospitalization receiving RBCs by admitting surgical service.
Volume **, ** **
TRANSFUSION
7
SHEHATA ET AL.
still required to identify critical thresholds for transfusion of PLTs in the cardiac surgery population,25 as the proportion of admissions receiving PLT transfusion has increased and it is unclear whether PLT transfusion is associated with increased morbidity and mortality.26,27 There are a few limitations to this study. Studies using administrative data are limited as they cannot describe the indications for transfusion or the use of alternative agents to treat anemia to explain the trends and fluctuations in transfusion. Yet, regardless of the indications, the trends categorized by clinical service identify disciplines where indications need to be further delineated. We did not describe laboratory data or patient characteristics and these features may have impacted on the trends in transfusion such as triggers for transfusion or age, as transfusion increases with older age, and there may have been an increase in the age of captured population in our study over time.28,29 This is a single-institution study and although generalization may be limited, the study included three campuses of a university-affiliated hospital and 244,013 hospitalizations. In summary, a clearer understanding of the indications and laboratory variables used to guide transfusion in patients admitted, particularly under the general internal medicine, nephrology services, and the critical care service for RBCs (as the proportion of admissions transfused in critical care is high despite data demonstrating that lower transfusion thresholds can be used11) is required. The general internal medicine service accounts for a large number of hospital admissions, and limited data are available on the indications for transfusion. Identifying factors that lead to high proportions of patients transfused RBCs under the critical care service may aid in identifying blood conservation strategies to decrease exposure to blood in these populations. Similarly there is the same need to determine the rationale for plasma transfusion in critical care and cardiac surgery and for PLT transfusion in cardiac surgery. The application of administrative data to describe trends in transfusion utilization and the proportion of admissions transfused in this study may also be useful in defining the patient populations and characteristics associated with increased transfusion rates. ACKNOWLEDGMENT We thank Robin Ducharme for her assistance in the preparation of the manuscript. CONFLICT OF INTEREST The authors have disclosed no conflicts of interest.
REFERENCES 1. Chiavetta JA, Herst R, Freedman J, et al. A survey of red cell use in 45 hospitals in central Ontario, Canada. Transfusion 1996;36:699-706. 8
TRANSFUSION Volume **, ** **
2. Wells AW, Mounter PJ, Chapman CE, et al. Where does blood go? Prospective observational study of red cell transfusion in north England. BMJ 2002;325(7368):803. 3. Cobain TJ, Vamvakas EC, Wells A, et al. A survey of the demographics of blood use. Transfus Med 2007;17:1-15. 4. Stanworth SJ, Cockburn HA, Boralessa H, et al. Which groups of patients are transfused? A study of red cell usage in London and southeast England. Vox Sang 2002;83: 352-7. 5. Stanworth SJ, Walsh TS, Prescott RJ, et al.; Intensive Care Study of Coagulopathy (ISOC) investigators. A national study of plasma use in critical care: clinical indications, dose and effect on prothrombin time. Crit Care 2011;15: R108. 6. Walsh TS, Garrioch M, Maciver C, et al.; Audit of Transfusion in Intensive Care in Scotland Study Group. Red cell requirements for intensive care units adhering to evidence-based transfusion guidelines. Transfusion 2004; 44:1405-11. 7. Chatterjee S, Wetterslev J, Sharma A, et al. Association of blood transfusion with increased mortality in myocardial infarction: a meta-analysis and diversity-adjusted study sequential analysis. Arch Intern Med 2013;173:132-9. 8. Carson JL, Hébert PC. Here we go again—blood transfusion kills patients? Comment on “Association of blood transfusion with increased mortality in myocardial infarction: a meta-analysis and diversity-adjusted study sequential analysis.” JAMA Intern Med 2013;173:139-41. 9. Yang L, Stanworth S, Hopewell S, et al. Is fresh-frozen plasma clinically effective? An update of a systematic review of randomized controlled trials. Transfusion 2012; 52:1673-86; quiz 1673. 10. Corwin HL, Gettinger A, Pearl RG, et al. The CRIT Study: anemia and blood transfusion in the critically ill—current clinical practice in the United States. Crit Care Med 2004; 32:39-52. 11. Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999;340:409-17. 12. Hébert PC, Blajchman MA, Cook DJ, et al.; Transfusion Requirements in Critical Care Investigators for the Canadian Critical Care Trials Group. Do blood transfusions improve outcomes related to mechanical ventilation? Chest 2001;119:1850-7. 13. Vincent JL, Baron JF, Reinhart K, et al. Anemia and blood transfusion in critically ill patients. JAMA 2002;288:1499507. 14. Williamson LM, Devine DV. Challenges in the management of the blood supply. Lancet 2013;381(9880):1866-75. 15. Guralnik JM, Eisenstaedt RS, Ferrucci L, et al. Prevalence of anemia in persons 65 years and older in the United States: evidence for a high rate of unexplained anemia. Blood 2004;104:2263-8.
TRENDS IN BLOOD UTILIZATION
16. Kulnigg S, Gasche C. Systematic review: managing anaemia in Crohn’s disease. Aliment Pharmacol Ther 2006;24(1112):1507-23. 17. Okonko DO, Grzeslo A, Witkowski T, et al. Effect of intravenous iron sucrose on exercise tolerance in anemic and nonanemic patients with symptomatic chronic heart failure and iron deficiency FERRIC-HF: a randomized, controlled, observer-blinded trial. J Am Coll Cardiol 2008;51: 103-12. 18. Silverberg DS, Iaina A, Schwartz D, et al. Intravenous iron in heart failure: beyond targeting anemia. Curr Heart Fail Rep 2011;8:14-21. 19. Auerbach M, Goodnough LT, Picard D, et al. The role of intravenous iron in anemia management and transfusion avoidance. Transfusion 2008;48:988-1000. 20. Wallis JP, Wells AW, Chapman CE. Changing indications
23. Estcourt L, Stanworth S, Doree C, et al. Prophylactic platelet transfusion for prevention of bleeding in patients with haematological disorders after chemotherapy and stem cell transplantation. Cochrane Database Syst Rev 2012;(5): CD004269. 24. Wandt H, Schaefer-Eckart K, Wendelin K, et al.; Study Alliance Leukemia. Therapeutic platelet transfusion versus routine prophylactic transfusion in patients with haematological malignancies: an open-label, multicentre, randomised study. Lancet 2012;380(9850):1309-16. 25. Zaffar N, Joseph A, Mazer CD, et al. The rationale for platelet transfusion during cardiopulmonary bypass: an observational study. Can J Anaesth 2013;60:345-54. 26. Spiess BD, Royston D, Levy JH, et al. Platelet transfusions during coronary artery bypass graft surgery are associated with serious adverse outcomes. Transfusion 2004;44:
for red cell transfusion from 2000 to 2004 in the North of England. Transfus Med 2006;16:411-7.
1143-8. 27. Karkouti K, Wijeysundera DN, Yau TM, et al. Platelet trans-
21. Tay J, Tinmouth A, Fergusson D, et al. Transfusion of red cells in hematopoietic stem cell transplantation (TRIST): study protocol for a randomized controlled trial. Trials
fusions are not associated with increased morbidity or mortality in cardiac surgery. Can J Anaesth 2006;53:279-87. 28. Drackley A, Newbold KB, Paez A, et al. Forecasting Ontar-
2011;12:207. 22. Stanworth SJ, Estcourt LJ, Powter G, et al.; TOPPS Investigators. A no-prophylaxis platelet-transfusion strategy for hematologic cancers. N Engl J Med 2013;368: 1771-80.
io’s blood supply and demand. Transfusion 2012;52:36674. 29. Greinacher A, Fendrich K, Brzenska R, et al. Implications of demographics on future blood supply: a population-based cross-sectional study. Transfusion 2011;51:702-9.
Volume **, ** **
TRANSFUSION
9
